News about Kimi K3.1 Pro

Chinese AI firm Moonshot AI launches Kimi K3.1 Pro, scoring 55 on the Artificial Analysis Intelligence Index — 4th globally, just 3 points behind Claude Opus 5.5. Massive 11-point jump over K3, driven by post-training compute. Leads in agentic tasks (GDPval-AA 1846 Elo), coding (Terminal-Bench 65.7%), 3D modeling (BenchCAD 92.7%), and office tasks. Same 2.8T params, 33% cheaper API ($2/M in, $10/M out). Weights planned for open release by Nov 2026 — would make it the top open-weights model, cutting the open-vs-prop gap from 12 to 3 pts. Cyber data stripped for safety. #MoonshotAI #KimiK3 #OpenWeights #AIbenchmark #LLM

Moonshot AI targets $2B annualized revenue by year-end, driven by its open-weight K3 model generating 300B tokens daily. Despite lower margins vs closed-weight rivals, the lab proves open-weight AI can still monetize—though it faces controversy over allegedly distilling Anthropic's Claude for K3 training. #MoonshotAI #OpenWeightAI #AIMarket #RevenueGrowth #TechNews

Moonshot AI targets $2B annualized revenue by year-end with its open-weight K3 model, doubling its August run rate. The goal highlights the financial potential of open-weight AI, though margins remain lower than closed-weight competitors. Meanwhile, Anthropic accused the lab of large-scale model distillation, routing nearly 300K requests to Claude Opus and collecting over 23 million responses for training. #MoonshotAI #OpenWeightAI #ModelDistillation #AIReward #K3

Moonshot AI open-sources Kimi K3 weights and infrastructure, claiming 2.5x compute efficiency. The Chinese model scores near Western frontiers like GPT-5.6 Sol at lower cost, but UK Cyber Institute tests reveal significant gaps in cyber and math capabilities — fueling distillation suspicions. While Chinese models often face such accusations, US open-weight advocates increasingly view distillation as legitimate. Kimi K3's release intensifies the open vs. closed frontier race. #MoonshotAI #KimiK3 #OpenWeights #AIModels #Distillation

New benchmark PerceptionBench reveals that even top multimodal AI models fail at basic visual perception, scoring under 60%. Many errors labeled as 'reasoning failures' are actually perception flaws. The test breaks vision into 10 atomic skills, showing limited progress across the board. #AI #MachineLearning #VisualPerception #Benchmark #MoonshotAI

Moonshot AI releases Kimi K2.6, an open-weight model rivaling GPT-5.4 and Claude Opus 4.6 on coding and agent benchmarks. It features a 300-agent swarm, 4,000+ tool calls, and 12+ hour runtime. Available via API, Hugging Face, and kimi.com under a modified MIT license with attribution requirements for large-scale commercial use. #KimiK26 #OpenWeight #AIAgents #MoonshotAI #CodingAI

Moonshot AI open-sourced Kimi K3 model weights and infrastructure (attention kernels, MoE library, agent tools), claiming 2.5x compute efficiency. Benchmarks show near-frontier performance at lower cost, but UK Cyber Institute found significant gaps in cyber and math capabilities vs. top models, raising distillation questions. Chinese models often face such accusations, though US open-weight advocates increasingly accept distillation as legitimate. #MoonshotAI #KimiK3 #OpenWeights #AIModels #Distillation

Moonshot pauses new Kimi K3 subscriptions after GPU capacity maxed out in 48 hours. Current users unaffected; new slots reopening gradually. Startup splits plans into "Kimi Membership" (general features) and "Kimi Code Membership" (coding workflows) to balance compute load. Meanwhile, Alibaba launches open-weight Qwen 3.8 with discounted preview access. #AI #KimiK3 #MoonshotAI #GPUShortage #Qwen3

Moonshot's Kimi K3 tops the Code Arena Frontend benchmark, surpassing Claude Fable 5 and GPT-5.6 Sol, becoming the first Chinese model to lead. However, it scores only ~39% on FrontierMath Tier 4, far behind OpenAI and Anthropic models (~90%). Mixed performance: strong in frontend coding, weak in complex math. #AI #KimiK3 #FrontendDevelopment #MathBenchmarks #MoonshotAI

Chinese startup Moonshot AI released Kimi K3, an open-source model claiming frontier-level performance rivaling top proprietary systems. Independent benchmarks confirm competitiveness, triggering market jitters — Nasdaq fell 1% amid Nvidia sell-offs. The launch, timed with Xi Jinping's AI speech, reignites debates echoing DeepSeek's R1 release, now amplified by Trump's tariff war, Anthropic security disputes, and looming AI IPOs. David Sacks blames US regulatory knots for losing the AI race; Travis Kalanick accuses China of model distillation. OpenAI's Dean Ball praises Kimi but warns open weights lead to "AI communism," predicting the Trump admin will weaponize regulatory FUD against Chinese models. Transformer's Shakeel Hashim counters that risks are overblown — Kimi lacks dangerous cyber capabilities, and China will face similar restraint incentives. #AI #MoonshotAI #KimiK3 #OpenSourceAI #Geopolitics

Chinese AI lab Moonshot prepares to launch Kimi K3, a massive 2–3 trillion parameter open-weight model expected to rival Anthropic's Opus 4.8. Release imminent as Moonshot seeks $31.5B valuation. Growing enterprise shift toward open-source alternatives like Moonshot, DeepSeek, and Z.ai challenges closed-source AI dominance. #MoonshotAI #KimiK3 #OpenSourceAI #AIModels #ChinaTech

Moonshot AI released Kimi K2.7 Code, an open-weights coding model for complex programming and agentic workflows. It trails GPT-5.5 and Claude Opus 4.8 on many coding benchmarks, but performs strongly in real-world agent tests and costs far less: $0.95 input/$4 output per million tokens. With 1T MoE parameters, 256k context and a modified MIT license, it may be a practical budget pick. #AI #OpenSource #Coding #MoonshotAI #LLMs

Moonshot AI released Kimi K2.7 Code, an open-weights coding model for complex programming and agentic workflows. It trails GPT-5.5 and Claude Opus 4.8 on many coding benchmarks, but performs strongly in real-world agent tests and costs far less: $0.95 input/$4 output per million tokens. With 1T MoE parameters, 256k context and a modified MIT license, it may be a practical budget pick. #AI #OpenSource #Coding #MoonshotAI #LLMs

Moonshot AI launched Kimi K2.7 Code, an open-weights model for complex coding and agent workflows. It improves on K2.6 and performs well in practical agent tests, though it still trails GPT-5.5 and Claude on many coding benchmarks. With a 1T-parameter MoE design, 256k context, multimodal support and API pricing of $0.95/$4 per million tokens, it may be a strong budget alternative. #AI #OpenSource #CodingAI #LLM #MoonshotAI

Moonshot AI, maker of Kimi chatbot, is pursuing a $30 billion valuation in a new funding round—over six times its late‑2025 worth—and aims to raise $1–2 billion. ARR has doubled to ~$200 million, its models rival GPT‑5.4 and Claude Opus 4.6, but it faces stiff competition from DeepSeek, Alibaba and MiniMax while preparing a possible Hong Kong IPO. #MoonshotAI #AIvaluation #FundingRound #HKIPO #ChineseAI

Moonshot AI aims for a $30B valuation in a new funding round, seeking $1‑2B after its revenue doubled to ~$200M. Its Kimi K2.6 model rivals GPT‑5.4 and Claude Opus 4.6, but faces stiff competition from DeepSeek, Alibaba and MiniMax, while preparing a possible Hong Kong IPO. #MoonshotAI #AIvaluation #Kimi #FundingRound #IPO

Moonshot AI releases open-weight Kimi K2.6, matching GPT‑5.4 and Claude Opus 4.6 on coding benchmarks. It supports up to 300 agents in parallel, chains >4k tool calls, runs >12 hrs, and features Agent Swarm for coordinated sub‑agents that can build full‑stack apps, websites, and documents. Available via chat, API, Hugging Face. #KimiK2_6 #AgentSwarm #OpenWeight #AIBenchmarks #MoonshotAI

Moonshot AI releases open-weight Kimi K2.6, rivaling GPT-5.4 and Claude Opus 4.6 on coding benchmarks. Supports up to 300 parallel agents, chains >4k tool calls, runs >12 hrs, enables full‑stack tasks via agent swarm. MIT‑licensed (attribution required for >100M users or $20M/month revenue). Available on kimi.com, API, Hugging Face. #KimiK2.6 #AgentSwarm #OpenWeight #AIModels #MoonshotAI